The growth of the HIV epidemic has been relentless, taking a devastating toll in terms of human health and social welfare. Setbacks in prevention interventions, including HIV vaccines and other behavioral and biomedical interventions, suggest that identifying effective prevention strategies may prove even more elusive than anticipated. In theory, one might conduct randomized clinical trials (RCTs) to estimate the efficacy of various promising intervention strategies on reducing HIV incidence among individuals and populations. However, for ethical and logistical reasons this is not feasible. One innovative alternative to long-term, randomized clinical trials is the use of stochastic modeling to allow for the simulation of a variety of interventions, alone and in combination, to assess their impact on HIV incidence and prevalence under a wide array of plausible assumptions. This proposal has the following three specific aims: AIM1: To build a stochastic, agent-based HIV transmission model which captures the population effects of transmission dynamics of HIV by incorporating sexual behaviors within a population that affect HIV transmission, linking the transmission model to the highly successful CEPAC US and CEPAC International models to capture the effects of HIV disease state and infectivity as a function of HIV RNA, treatment resistance, and adherence on HIV transmission in resource limited settings and the US and to calibrate the model to data from both the Republic of South Africa and the US before ART roll out. AIM 2: In the Republic of South Africa, to evaluate the impact of primary, secondary and combination HIV prevention interventions on population level measures of effect (HIV incidence, prevalence, life expectancy), cost and cost-effectiveness) in the epidemic. AIM 3: In the US, to evaluate the impact of primary, secondary and combination HIV prevention interventions either alone or in combination, on population level measures of effect (HIV incidence, prevalence, quality adjusted life expectancy), cost and cost-effectiveness among MSMs.